When Words Exclude: Newspaper Sentiment and the Economic Consequences of the Chinese Exclusion Act

• In Event: Inequality, Poverty, and Mobility Section Roundtables
  In Refereed Roundtable (60 minute): 17. Heterogeneity

Abstract

This study examines how local newspaper discourse shaped migration and labor market outcomes during the era of the Chinese Exclusion Act of 1882 in the United States. While existing research has focused primarily on the legal and aggregate economic consequences of exclusionary immigration policy, less attention has been paid to the role of local social attitudes and media narratives in influencing migration behavior and labor market adjustment. I argue that newspapers were a key channel through which anti-Chinese sentiment was formed, reinforced, and transmitted at the local level, shaping both individual decisions and broader economic dynamics. Extracting thousands of digitized newspapers from Chronicling America and Newspaper.com, I construct a novel county-level measure of anti-Chinese sentiment based on the frequency, intensity, and context of exclusionary language in the years preceding the Act (1870-1874). In addition to keyword frequency, I apply machine learning and natural language processing techniques, specifically Sentence-BERT (SBERT), to capture semantic similarity to exclusionary rhetoric. Newspaper mentions are further weighted by linking historical directory data to account for circulation and local exposure. Exploiting cross-county variation in pre-Exclusion sentiment, I estimate how exposure to hostile media environments influenced the likelihood of migration, distance moved, and subsequent labor market trajectories. I also examine spillover effects on non-Chinese workers and local labor markets. By combining large-scale text analysis with linked historical microdata, this study provides new evidence on the economic consequences of ethnic discrimination and highlights the causal role of local media in shaping migration and labor market outcomes.

Author

• Olivia Finan, University College Dublin